Towards Deep Learning Models Resistant to Adversarial Attacks
@article{madry2017towards,
title={Towards Deep Learning Models Resistant to Adversarial Attacks.},
author={Madry, Aleksander and Makelov, Aleksandar and Schmidt, Ludwig and Tsipras, Dimitris and Vladu, Adrian},
journal={arXiv: Machine Learning},
year={2017}}
概
利用特定的方法产生"坏"样本(Adversarial samples), 以此来促进网络的稳定性是当下的热点之一, 本文以实验为主, 比较PGD( projected gradient descent) 和 FGSM(fast gradient sign method)在不同数据下的表现, 以及由普通样本产生"坏"样本会出现的一些现象.
主要内容
Adversarial attacks 主要聚焦于下列问题:
\min_{\theta} \rho (\theta) \quad where \quad \rho(\theta) =\mathbb{E}_{(x,y)\sim D}[\max_{\delta \in S} L(\theta, x+\delta, y)].
\]
其中\(S\)是我们指定的摄动集合, 直接一点就是\(|\delta| <constant\)之类.
通过FGSM产生"坏"样本:
\]
这个思想是很直接的(从线性感知器谈起, 具体看here).
PGD的思路是, 给定摄动集\(S\), 比如小于某个常数的摄动(e.g. \(\{\tilde{x}:\|x-\tilde{x}\|_{\infty}<c\}\)), 多次迭代寻找合适的adversarial samples:
\]
其中\(\prod\)表示投影算子, 假设\(S=\{\tilde{x}:\|x-\tilde{x}\|_{\infty}<c\}\),
\]
实际上, 可以分开讨论第\((i,j)\)个元素, \(y:=(x^t + \alpha \: \mathrm{sgn} (\nabla_x L(\theta,x, y))\), 只需找到\(z_{ij}\)使得
\]
最小即可. 此时有显示解为:
\left \{
\begin{array}{ll}
x_{ij} +c & y_{ij} > x_{ij}+c \\
x_{ij} -c & y_{ij} < x_{ij}-c \\
y_{ij} & else.
\end{array} \right.
\]
简而言之就是一个截断.
重复几次, 至到\(x^t\)被判断的类别与初始的\(x\)不同或者达到最大迭代次数.
Note
- 如果我们训练网络能够免疫PGD的攻击, 那么其也能很大一部分其它的攻击.
- FGSM对抗训练不能提高网络的稳定性(在摄动较大的时候).
- weak models may fail to learn non-trival classfiers.
- 网络越强(参数等程度)训练出来的稳定性越好, 同时可转移(指adversarial samples 在多个网络中被误判)会变差.
Towards Deep Learning Models Resistant to Adversarial Attacks的更多相关文章
- How to Grid Search Hyperparameters for Deep Learning Models in Python With Keras
Hyperparameter optimization is a big part of deep learning. The reason is that neural networks are n ...
- a Javascript library for training Deep Learning models
w强化算法和数学,来迎接机器学习.神经网络. http://cs.stanford.edu/people/karpathy/convnetjs/ ConvNetJS is a Javascript l ...
- Run Your Tensorflow Deep Learning Models on Google AI
People commonly tend to put much effort on hyperparameter tuning and training while using Tensoflow& ...
- What are some good books/papers for learning deep learning?
What's the most effective way to get started with deep learning? 29 Answers Yoshua Bengio, ...
- (转) Awesome Deep Learning
Awesome Deep Learning Table of Contents Free Online Books Courses Videos and Lectures Papers Tutori ...
- (转)分布式深度学习系统构建 简介 Distributed Deep Learning
HOME ABOUT CONTACT SUBSCRIBE VIA RSS DEEP LEARNING FOR ENTERPRISE Distributed Deep Learning, Part ...
- The Brain vs Deep Learning Part I: Computational Complexity — Or Why the Singularity Is Nowhere Near
The Brain vs Deep Learning Part I: Computational Complexity — Or Why the Singularity Is Nowhere Near ...
- Paper Reading——LEMNA:Explaining Deep Learning based Security Applications
Motivation: The lack of transparency of the deep learning models creates key barriers to establishi ...
- Coursera Deep Learning 2 Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization - week1, Assignment(Regularization)
声明:所有内容来自coursera,作为个人学习笔记记录在这里. Regularization Welcome to the second assignment of this week. Deep ...
随机推荐
- Hive(十三)【Hive on Spark 部署搭建】
Hive on Spark 官网详情:https://cwiki.apache.org//confluence/display/Hive/Hive+on+Spark:+Getting+Started ...
- Scala(五)【集合的高级使用】
目录 一.集合属性 size length contains mkString 二.集合方法 drop.dropRight distinct:去重 head.last:获取第一个.最后一个元素 tai ...
- Sharding-JDBC 实现垂直分库水平分表
1.需求分析
- Shell学习(二)——变量和基本数据类型
参考博客: [1]LinuxShell脚本--变量和数据类型 [2]shell只读变量删除 一.变量 定义变量的语法 定义变量时,变量名和变量值之间使用"="分隔,并且等号两边不能 ...
- Android 极光推送集成
集成Jpush 1.用Android Studio创建一个Demo 2.创建激光推送开发者账号,要创建极光推送开发者帐号,请访问极光推送官方网站https://www.jiguang.cn/push ...
- Oracle 创建 md5 加密函数
使用 Oracle 的 utl_raw.DBMS_OBFUSCATION_TOOLKIT 可以获取 md5 加密字符串: select utl_raw.cast_to_raw(DBMS_OBFUSCA ...
- spring基于注解的声明式事务控制
package com.hope.service.impl;import com.hope.dao.IAccountDao;import com.hope.domain.Account;import ...
- 【C/C++】习题3-7 DNA/算法竞赛入门经典/数组与字符串
[题目] 输入m组n长的DNA序列,要求找出和其他Hamming距离最小的那个序列,求其与其他的Hamming距离总和. 如果有多个序列,求字典序最小的. [注]这道题是我理解错误,不是找出输入的序列 ...
- 前端避坑指南丨辛辛苦苦开发的 APP 竟然被判定为简单网页打包?
传统混合移动App开发模式,通常会使用WebView作为桥接层,但随着iOS和Android应用商店审核政策日趋严格,有时会被错误判定为简单网页打包成App,上架容易遭到拒绝. 既然可能存在风险,那我 ...
- vue文件上传及压缩(canvas实现压缩)
// 读取文件结果 afterRead(files) { let that = this; let file = files.file; if (file === undefined) { retur ...